The Status of Grid Computing North Carolina Statewide Grid NC State University, Jan 20, 2005 Wolfgang Gentzsch Managing Director MCNC, Grid Computing and Networking Services, and Department of Computer Science at NC State University This Lecture aims at complementing your knowledge in distributed computing technologies and developments, providing an actual overview of Grid Computing, and its applications and benefits for Education, Industry and Economy. A simple View of Grid : An IT Utility on a . . . Grid Middleware (the glue) Managing . . . Networked Distributed Resources Integrated & “Integrate-able” End-to-End, Javacard -> J2EE Innovation The Grid Partners The underpinning technology for distributed computing Flexibility Security Scale Enables coordinated sharing of distributed and dynamic resources, including computers, applications, data archives, visualization, sensors, and multiple remote instruments Open public API's Value Why Should We Care about Grids ? “ It’s the next big thing ! “ Grid technologies advance Science and Education in that we can do things which haven’t been possible before. Grid infrastructure attracts and enables new businesses and creates new jobs, especially in today's rural areas. Grids make us more competitive by better utilizing resources, bringing results ‘to market’ faster, and delivering with higher quality. An early pilot: NC BioGrid One of the nation’s first grid test beds for computing, data storage and networking resources for life sciences research Established to research and implement new grid computing technologies that will enable researchers and educators throughout North Carolina to take full advantage of the genomic revolution • Installed in Summer 2002, heterogeneous . hardware and OS platforms • More than 80 organizations • Dedicated systems for testing grid middleware and developing grid applications for bioinformatics • Spans multiple administrative domains with systems located at MCNC, NC State, UNC-CH & Duke • Established a Certificate Authority A Real Grid: NEES, Network for Earthquake Engineering Simulation NEESgrid will link earthquake researchers across the U.S. with leading-edge computing resources and research equipment, allowing collaborative teams to plan, perform, and publish their experiments. Through the NEESgrid, researchers will: - perform tele-observation and tele-operation of experiments; - publish to and make use of a curated data repository using standardized markup; - access computational resources and open-source analytical tools; - access collaborative tools for experiment planning, execution, analysis, and publication. The Future just Began: SURA SCOOP South-eastern Coastal Ocean Observing Program The Challenge: - More than half of the nation's tidal shores home to 80 million people - coastal zone is undergoing environmental and ecological changes - threaten the sustainability of the region's economies and marine resources The Solution: - develop a Grid of sensors and linked computers - fully integrating several observing systems in the southern region - provide data, in real-time and at high speed, for more reliable, accurate and timely information To help guide effective coastal stewardship, plan for extreme events, facilitate safe maritime operations, and support coastal military security. Customer Value Proposition (Benefits) Department, Enterprise, and Global Grids • • • • • • • • • Access: transparent, remote, secure, wireless Sharing: enable collaboration over the network Failover: migrate/restart applications automatically On Demand: get resources, when you need them Productivity: more work done in shorter time Virtualization: access compute services, not servers Heterogeneity: platforms, OSs, devices, software Resource Utilization: increase from 20% to 80+% Virtual Organizations: build & dismantle on the fly However, There is Still a Long Way to Go ! Grids are over-hyped: currently, they promise much more than they can really offer. Grid technology is far from mature and complete. Grid standards are (mostly) still missing. Grids are very complex IT infrastructures. Grids bring new challenges: sharing resources, loosing direct control, security, intellectual property, legal, social, political issues . . . Grid Architecture, Technology, Standards Architectural Requirements are huge • Availability • • • Downtime impact Individual jobs Maintenance windows • Scalability • • • • Growth 1-3-5 years Scaling strategy and Response to peak loads Technology refresh, evolution • Manageability • • • Skill set / workload of SA Expected stability Code management, software distribution • Security • • • User authentication Internet access Data Security requirements • Data Distribution • • Location, volume, refresh Security of data • Usability • • • Administrative Skill set and Client environment Psychological factors important • Operations Management • • • 100s CPUs / SA Resources added in large blocks Change control is critical The Grid Landscape is Changing Developer View: Write here Run anywhere Service Composition HPTC Sys Mgmt Standards Public SOA Web Svcs Utility Computing OnDemand “Compute Power from the wall socket” Outsourcing/Hosting Software as a Svc ... Technologies HP IBM IBM Sun Enterprise IT Mgmt Products Details see backup slides Market making / Productization Enabled Business Models Courtesy: John Tollefsrud, Sun Microsystems 2001: Global Grid Forum (GGF) • Community-driven set of working groups that are developing standards and best practices for distributed computing ("Grids“) efforts • 2001: Merger of US, APAC, Euro Grid Forums • 2002: Open Grid Services Architecture (OGSA) • 2004: WS-RF Web Services Resource Framework • Standards: IETF, DMTF, OASIS, WS-I, W3C, EGA,… 2002: Open Grid Services Architecture OGSA Grid Technologies Web Services OGSA Open Grid Service Architecture ● Integrates grid technologies with Web Services (OGSA => WS-RF) ● Defines the key components of the grid ● 2005: Globus Toolkit 4.x Open Grid Services Architecture (OGSA) Web Services Resource Framework (WSRF) • • • Integration of Grid technologies and Web Services OGSA defines a “Grid Service” In terms of WSDL interfaces, defines mechanisms: creating distributed systems lifetime management change management ... other interfaces Grid Service Service Data Element Service Data Element Service Data Element Implementation / Interfaces Hosting Env/runtime: C, J2EE Systems Infrastructure credential management notification Notification Authorization Service Creation Service Registry Manageability Concurrency NCGrid Globus Toolkit 4 Workshops Friday, January 28th, and Friday, February 4th, 2005 Presenter: Pawel Plaszczak, Gridwise Technologies; Contact: Chuck Kesler jckesler@mcnc.org Day 1 - Friday, January 28th Morning Session (9am - noon) · Introduction to grid computing and the Globus Toolkit (brief) · Review of GT 2.4, 3.2, 4.0 and counterpart standards (OGSA, WSRF) · Overview of Globus components, including topics not covered (e.g. OGSA-DAI, CSF, CAS) · Discussion of GSI security (X.509 certs, chain of trust, CA, signing certificates, proxies, secure WS components) Afternoon Session (1pm - 4pm) · Web services basics · Intro to Java tools for Grid/Web Service/WSRF programming (Log4j, Ant, Apache Axis) · GT4 Java WS core (WSRF) · Overview of the WSRF standard · Overview of GT4 Grid service programmer environment (schema, WSDL, compiling, stub generation, deployment, testing) Day 2 - Friday, February 4th Morning Session (9am - noon) · GRAM and job submission / purpose and architecture overview · Web Services GRAM vs. pre-Web Services GRAM · Adapting to customer scheduler interfaces (e.g. SGE, LSF, PBS) · Information Services (MDS4) · GridFTP, RLS, XIO (theory) Afternoon Session (1pm - 4pm) · Hands-on workshop - details to be announced Grid Computing Today in Industry Mid 90s: Cluster Grids Sun Fire V100 server Login host SGE submit host Sun Fire 280R server SGE master SunMC server Sun Blade 100 system SGE admin host SunMC console Sun Fire 280R server Large SMP Management tier NFS server Install server Backup server Linux system Sun Fire V880 server Access tier Compute tier with 2 CPUs Linux system with 2 CPUs Sun Fire V880 server Large SMP Linux system with 2 CPUs Linux system Sun Fire 6800 server Large SMP with 2 CPUs Linux system with 2 CPUs Myrinet Late 90s: Department Grid Browser to CGM (Remote Server Setup & Configuration) I n t e r n e t Compute Grid Manager Auto OS Deployment Grid Installation/Mgmt Centralized Server Mgmt Solaris Servers Linux Servers Workstations (Linux or Solaris) 2001: Enterprise Grid Reference Architecture Browser Access via GEP SunRay Access Workstation Access Optional Control Network (Gbit-E) Myrinet Myrinet Servers, Blades, & VIZ Myrinet Linux Racks Grid Manager Workstations Sun Fire Link Data Network (Gbit-E) Gbit-E switch V240 / V880 NFS V240 / V880 NFS Gbit-E switch V240 / V880 NFS Gbit-E switch V240 / V880 NFS Gbit-E switch Gbit-E switch V880 QFS/NFS Server V880 QFS/NFS Server FC Switch NAS/NFS Simple NFS HA NFS Scalable QFS/NFS 2002: Globus & Avaki over mult Department Grids 1 2 SGE cluster mgmnt within an admin domain & file system area Globus/Avaki connects resources, handles files, binary management, and high level resource selection Organization A Organization B a d Proprietary Data b e c SGE f g SGE SGE Data mapped and available through Avaki Data Grid Clients view Organization C Avaki HPC / a b d e f c g h Avaki Data Grid h GOAL: Computing as a Utility =>> Enhancing The Grid with a Business Model What's a Utility? • On Demand: Get a service at your finger tip From the Wall Socket: Don't care about the infrastructure • Metering & Billing: Pay as you go, for what you use • Like electricity, water, gas, heat, telephony Who Uses Grids today? • Life Sciences ● Startup and cost efficient Electronic Design Time to Market Custom research or limited use applications Multi-day application runs (BLAST) Fastest platforms, largest Grids Exponential Combinations License Management Limited administrative staff Complementary techniques Well established application suite Large legacy investment ● Financial Services Market simulations Time IS Money Proprietary applications Multiple Platforms Multiple scenario execution Need instant results & analysis tools Platform Ownership issues ● High Performance Computing Parallel Reservoir Simulations Geophysical Ray Tracing Custom in-house codes Large scale, multi-platform execution MCNC: 3-Year GCNS Grid Roadmap Grid Service Provider: Awareness Creation: Network, Computing, Data, Video Partner with SC Sites Build GTEC Service Portfolio Start Grid Consulting Easy Access: Annual GSP Workshop Training, Web Courses Access Grid Node Grid Appliance QuadA Conferences, Workshops, PR MCNC Enterprise Grid NC BioGrid NC Statewide Grid CY03 CY04 CY05 CY06 North Carolina’s Foundation for Grid: NCREN Internet Internet 4-7 MCNC-owned Clusters 2 distributed throughout the state NLR Locations still under evaluation Cisco Interne t Existing: Blend of owned and leased fiber and circuits moving toward resilient rings powered by Cisco routers Planned: Strong focus on owned and leased fiber, Lambda, and few circuits, in resilient rings powered by Cisco routers and Wave Division Multiplexers EPA GCNS Enterprise Grid Avaki Data Grid Global Grid Resource DB (GIIS) Users Campus Grids Data Grid Access Servers (8 total, i.e. 1 per 8 nodes) Grid Gatekeeper / Interactive Nodes LSF Master Job Scheduler 32-CPU SGI Altix Linux SMP Server 128-CPU IBM Linux Cluster (64 nodes) 8-TB Storage Network, Grid and Data Center Services GTEC, NLR, ANR and other Innovation Initiatives DEPLOYMENT Self-serve Data Center Services Value-add Information Systems Services Hosting & Infrastructure Enterprise Grid Services State-wide Grid Services Grid Computing DATA CENTER NCREN Data Archival Services Information Security Services Information Assurance The NC State-Wide Grid Roadmap • • • • • • • • • • • 06/04: “Do-Grid-Yourself” Workshop 07/04: Phase 1, Awareness Creation 09/04: WCU’s Barry Wilkinson Grid Lectures 10/04 – 06/05: Deliver/Connect Grid Appliance Clusters to University Partners in NC 10/04: Develop “Do-Grid-Yourself” Training Course 10/04: Start QuadA Project: Access, Accounting, Authentication, Authorization 12/04: Start Deliver Grid Training to Partner Univs 02/05: Globus Toolkit Training 01/05 – 06/05: Work with Grid Users to Port Apps 03/05: Build Access Grid Node 06/05: 1st NC Statewide Grid Workshop Example: The North Carolina Start-up Grid With NC State University & Centennial Campus Incubator - To help Start-ups overcome manifold obstacles - To reduce Start-up risk and increase success - To attract Start-ups to North Carolina - To contribute to North Carolina’s economic growth Easy Start: Leverage what Exists • All-Inclusive Office Space: Centennial Campus Incubator exists • IT Resources: MCNC’s Enterprise Grid and NC Statewide Grid exist • Grid Access: Grid Appliance and Grid Portal prototype exist • Web Presence: For company product information and marketing • Business Platform: To offer any (grid-enabled) digital service to users, customers, and partners, over the Web/Grid • Consulting: From NC State (e.g. College of Mgmnt) and the Triangle Grid Vision Our Vision : The Three Waves of Grid Computing The Research Wave Technology, Prototypes Virtual Organizations Standards GGF, IETF, OASIS GCNS: “Awareness Creation” The Industry Wave Grid-Enabled Products Enterprise Solutions Interoperability GGF, EGA, IETF, OASIS GCNS: “Easy Access” Commodity IT Utility Integration Legal, Ethical, Political Orgs GCNS: “Grid Service Provider” The Consumer Wave Grids Today - Grids in 3 - 5 Years Focus on Research Compute-oriented - Focus on R&D and business - Petaflops linked w/ Petabytes Computing Services - Web Services Proprietary interfaces - Standards: GGF, OGSA, WS-RF, DRMAA “Mental Firewall” - Security, policies, identity Difficult to build - Standards, services, solutions Difficult to manage - Sun N1, IBM Autonomic, HP DC,… Difficult to implement & use - Grid Portals: transparent, remote, secure Many technologies - Globus Toolkit 5.x, ??? Finally: Anyone, anywhere, anytime, any device, any data, connected to The Grid Integration of new devices, data and information sources Cell phones, PDAs, smart sensors, sensor arrays, health monitors Devices embedded in cars, engines, roads, bridges, clothes,... Huge amount of data for real-time analysis Policies, grid economy, to maintain stability and efficiency Support organizational and societal structures, to bridge political and social boundaries . . . very much like other vital infrastructures (roads, telecom,…) Time Machines The Innovation Engine Thank You ! wgentzsch@mcnc.org http://www.mcnc.org